Error: Unused Output Layer

20 次查看（过去 30 天）
Elaine Todd
Elaine Todd 2020-1-28
回答： Elaine Todd 2020-1-31
This is the full error I'm getting: Error using trainNetwork (line 150)
Invalid network.
Error in train (line 12)
Caused by:
Network: Missing output layer. The network must have one output layer.
Layer 4: Unused output. Each layer output must be connected to the input of another layer.
X is a 5 by 30 matrix that I'm trying to run through a relu, then a fully connected layer, and then a softmax to label the data as 0 or 1. I'm confused on why the softmaxLayer isn't taking in the data from the fullyConnectedLayer.
This is the function I'm trying to run.
function [trained_net] = train(X,Y)
% creates adam optimizer
options = trainingOptions('adam', 'InitialLearnRate',3e-4, ...
'SquaredGradientDecayFactor',0.99, 'MaxEpochs',20, ...
'MiniBatchSize',10, 'Plots','training-progress');
% creates layers
layers = [sequenceInputLayer(5)
reluLayer
fullyConnectedLayer(1)
softmaxLayer];
% trains network
trained_net = trainNetwork(X,Y,layers,options);

请先登录，再回答此问题。

回答（2 个）

Elaine Todd
Elaine Todd 2020-1-31
Turns out that even though Softmax is in the output layer category, it isn't counted as an output. I added in a classification output layer after the softmax and it runs now.
Harsha Priya Daggubati
Hi,
I guess the issue is with fullConnectedLayer's outputSize, as it needs to be equal to the number of classes your network needs to classify. Try changing it to 2 and check whether it works.

类别

在 Help Center 和 File Exchange 中查找有关 Image Data Workflows 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by